Automated threshold perimetry has increasingly become the basis for assessing the adequacy of glaucoma management, yet there are few data suggesting what constitutes normal variation in otherwise well controlled glaucoma, what constitutes a meaningful deviation in a patient with progressive optic nerve damage and how much damage occurs before such differences in the visual field can reliably detected. Automated perimetry generates a wealth of quantitative data, but analytic strategies which might best detect meaningful changes in the visual field have yet to be developed and evaluated. The assessment of progressive visual field loss is of practical importance not only to clinicians in the management of individual patients, but also for evaluating the efficacy of glaucoma treatment in clinical trial settings. This study proposes to develop analytic strategies for assessing progression of visual field loss and to evaluate them against other measures of optic nerve damage. Consecutive automated visual fields of control subjects, patients with ocular hypertensive and glaucoma will be assembled from a variety of sources.The temporal variability of visual field testing will be estimated for each of these groups. Covariates which affect this variability will be used to better distinguish meaningful change from chance variation. Estimates of visual field change will be obtained by regressing several measures of diffuse and localized loss against time and comparing these to other measures of optic nerve damage such as progressive cupping, nerve fiber layer deterioration and changes in manual perimetry. Individual slope estimates will also be improved by adjusting the estimates with data from other subjects at a similar stage of disease.
Showing the most recent 10 out of 13 publications
